In 165 healthy children and adolescents ranging from 7 to 18 years of age, we examined how age and sex affect exercise performance with peak and submaximal CPET parameters and two-dimensional analysis. With this combined approach, the age-related exercise performance may mainly be characterized by a) age-related increase of body size (height and weight), b) sex-dependent changes due to differences in body composition, and c) sex-independent functional maturation.
Enhanced Exercise Performance with Increased Physical Growth
Rapid growth of body size or growth spurt commonly occurs during adolescence consisting of increase in bone length and mass, skeletal muscle mass, heart size, intravascular volume, and lung volume, all of which are known to contribute to increased pVO2 and exercise performance [8, 18]. Strong positive linear relationships between weight and pVO2 in both males and females support this phenomenon (Fig. 1) [19]. In our cohort, males showed progressive physical growth even after age 15 years, whereas females did not, as shown in Table 1, indicating a growth spurt in females may complete earlier than in males. This is in parallel with peak CPET parameters which still increase beyond age 16 years in males, whereas there is no significant difference between the 12 to 15 yo and ≥ 16 yo subgroups in females. DVO2/DHR and pOP, both surrogate parameters of stroke volume during submaximal and at peak exercise, respectively, increased in proportion to the body growth (Table 1), suggesting corresponding heart size increase with physical growth [20, 21]. Peak VE also increased with physical growth, suggesting growth of lung volume and augmentation of respiratory muscle with growth spurt in both sexes. Increase of peak systolic BP, pVO2 (L/min), and pWR also occurred in concordance with the physical growth. These findings regarding age-related changes in exercise performance are in agreement with the previously published studies [8, 22].
Sex-Dependent Effects on Exercise Performance
In addition to physical growth-dependent increase in CPET parameters, there were considerable sex differences in exercise performance more evident in older groups, especially in the ≥ 16 yo, suggesting the distinct effects of puberty on exercise performance by sex (Fig. 5). A growth spurt is induced by a growth hormone surge, followed by a burst of sex hormones, androgen and estrogen for males and females, respectively. Androgen generates strong anabolic effects as well as virilization that further increases bone and muscle mass, whereas estrogen enhances deposition of body fat in a special manner [23]. Additionally, heart size and stroke volume were noted to be higher in adolescent males [24] as well as lung volume and VE [25], thus allowing males to exhibit better exercise performance.
In Fig. 1, males demonstrated progressive increase in weight-pVO2 slope with age whereas there were no such age-related slope changes in female subgroups. Peak WR increased in accordance with physical growth in both sexes, but only males showed significantly higher pWR/kg in 12 to 15 yo and ≥ 16 yo than in ≤ 11 yo, suggesting not only increase in muscle mass but also augmented power production efficiency in older males. Enhanced skeletal muscle performance in older males is also supported by the age-related changes in regression lines between VAT and pVO2 (Fig. 3), indicating more sustainable muscle metabolism under an anaerobic condition beyond AT in older males [26]. This may be explained, in part, by higher glycogen utilization in skeletal muscles of males than females [27, 28]. This male puberty-induced enhanced skeletal muscle performance at peak exercise enables post-pubertal adolescent males to take on a greater pWR. Similar age-related enhanced ventilatory efficiency for oxygen uptake indicated by the pVE-pVO2 relationship was more prominent in males than in females (Fig. 4) with significant sex difference in ≥ 16 yo (Fig. 5). A larger chest cavity, larger lung volume, and stronger respiratory muscle in males may explain this phenomenon.
Another pattern identified was female age-dependent disadvantages in exercise performance likely due to their body composition. Despite significant increase of pVO2 in 12 to 15 yo and ≥ 16 yo females compared with that in ≤ 11 yo females, pVO2/kg was significantly lower in 12 to 15 yo and ≥ 16 yo females than in ≤ 11 yo females, suggesting the increased non-muscle mass or body fat mass in older females most likely due to female puberty. A similar trend was also noted in pOP/kg, VAT/kg, and D[VO2/kg]/DHR (Table 1). During female puberty, estrogen and progesterone levels rise significantly resulting in a large increase in body fat by approximately 6–10% in female adolescents [8]. When pVO2 is calculated relative to lean body mass instead of total body mass, these sex differences are reduced by approximately one half [29]. The difference in peak VO2 between males and females disappears when pVO2 is indexed by estimated leg muscle mass [30]. Overall differences in exercise performance between males and females ≥ 16 yo are likely attributed to both male pubertal advantages with higher body stature and considerably increased muscle mass, strength, and endurance and female pubertal disadvantages with higher body fat mass.
Sex-Independent Functional Maturation
We also identified age-related enhanced functional maturation in exercise performance independent of sex. This includes a significantly higher peak RER, lower ΔVO2/ΔWR, and lower ΔVE/ΔVCO2 in older subgroups than those in the ≤ 11 yo subgroups (Table 1), suggesting age-related substrate utilization in favor of carbohydrate metabolism, better work efficiency (or lower oxygen cost per work rate), and higher ventilatory efficiency, respectively [22]. Lower ΔVO2/ΔWR in 12 to 15 yo and ≥ 16 yo than in ≤ 11 yo in both sexes may reflect maturation of energy metabolism, suggesting more efficient kinetics of high-energy phosphate metabolism in muscle cells in older groups [31, 32]. However, low ΔVO2/ΔWR also represents decreased physical conditioning [33]. Independently, obese subjects tend to show elevated ΔVO2/ΔWR because of the excessive weight they carry [6]. The physiological significance of ΔVO2/ΔWR is multifactorial, which makes interpretation of lower ΔVO2/ΔWR in older adolescents challenging. Ventilatory response to exercise, represented by DVE/DCO2 slope, decreased with age because of changes in the control of ventilation that permits higher values of arterial CO2 partial pressure during exercise in adolescents [34]. Higher DVE/DCO2 implies either higher physiological dead space and/or increased ventilation-perfusion mismatch. One study indicates an element of advantageousness associated with motor learning rather than motor performance that is enhanced during adolescence [12]. This can be attributed to maturation of the central nervous system that is enhanced with age.
Two-Dimensional CPET Analysis
In this study, we used “two-dimensional CPET analysis”, unique and simple CPET interpretation methods, to aid understanding of exercise physiology [15, 16, 35]. In addition to interpreting each isolated CPET value, we assessed one CPET value in context with other parameters to understand the difference of general trends of the two groups. Figure 1 showed how body weight affects pVO2, demonstrating the degree of skeletal muscle effects, both quantitatively (muscle mass) and qualitatively (efficiency of muscle energy metabolism), at a given weight. There was an age-dependent increase in slope of the regression lines in males whereas there was no significant difference in females. Exercise endurance beyond AT was assessed by the relationship between VAT and pVO2 (Fig. 2), which presents upward shift of regression lines with age in males, suggesting male growth provides better endurance under anaerobic conditions [27]; this upward shift was relatively modest in females. The slope of regression lines between pVE and pVO2 in males demonstrated progressive upward and rightward shift with increase in age whereas such age-related differences were not as prominent in females. Data from this new approach are closely aligned with existing data of both peak and submaximal parameters [1, 4, 36].
Limitations
Our study has several important limitations. First, body composition was analyzed only through BMI, not by a direct measurement of lean body mass and/or body fat mass. The estimation of body composition only by BMI has an intrinsic limitation. Second, sex maturation was not individually assessed in our study. Presence of puberty or sexual maturation was only estimated by age. Onset, speed, and duration of puberty may also be variable among individuals, which were not assessed in this study. Third, the degree of physical conditioning was not controlled in this study. Fourth, although we used pOP and DVO2/DHR as surrogates of stroke volume, these values are also affected by hemoglobin levels and peripheral oxygen extraction, neither of which was measured in this study. Fifth, although our patients were those without underlying heart disease, they came to cardiology clinic for particular reasons. They may not necessarily represent ordinary healthy pediatric population. There is an intrinsic selection bias in patient enrollment. Last, a small cohort size limits the power of statistical analysis.
The first four limitations are fundamental problems at most of clinical settings, as important parameters, including lean body mass, assessment of pubertal maturation, and a level of physical conditioning, are not routinely available. Most frequently, standard reference values for CPET are provided only by absolute values or weight-indexed values. As a measured weight does not uniformly represent a muscle mass, weight indexed-CPET values are inevitably subject to certain error. Substantial gaps exist between the science of exercise physiology and clinical CPET in pediatric practice, especially through pubertal ages. We need to acknowledge these existing gaps and should put maximum efforts into overcoming the discrepancies. Despite these limitations, our study clearly characterized the changes in exercise performance during age-related growth and functional maturation by relatively simple methods that can be easily introduced in routine clinical practice for better understanding of exercise physiology and cardiopulmonary reserve.